#20459: ewma_cmp_cmux never considers policies different -------------------------------------------+------------------------------- Reporter: pastly | Owner: Type: defect | Status: needs_review Priority: High | Milestone: Tor: | 0.2.9.x-final Component: Core Tor/Tor | Version: Tor: | 0.2.6.2-alpha Severity: Normal | Resolution: Keywords: 029-backport, review-group-11 | Actual Points: Parent ID: | Points: Reviewer: | Sponsor: -------------------------------------------+-------------------------------
Comment (by pastly): Also FWIW, I'm currently running this change in small Shadow networks as I'm testing other things. I'm measuring kernel queue time, tor queue time, network throughput, and time to download. By these four metrics, this change doesn't hurt a relay in a small network. It also doesn't really seem to help. I started a large Shadow simulation a couple days ago, but it died sometime last night :(. I've started it again. I will report back when it has finished. I suspect the small network was too small for the difference to be noticeable. Namely - not enough channels from one relay to ~all others - not enough circuits per channel The large network has 500 relays and 7500 clients compared to the small's 11 and 12 respectively. -- Ticket URL: <https://troodi.torproject.org/projects/tor/ticket/20459#comment:11> Tor Bug Tracker & Wiki <https://trac.torproject.org/> The Tor Project: anonymity online _______________________________________________ tor-bugs mailing list tor-bugs@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs